InferenceIllusionist commited on
Commit
aa43199
1 Parent(s): aa164dd

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -2
README.md CHANGED
@@ -11,9 +11,14 @@ tags:
11
 
12
  <b>Highly requested model.</b> Quantized from fp16 with love.
13
 
14
- * 1st batch (IQ3_S, IQ3_XS) use a imatrix.dat file calculated from Q8 quant using an input file from [this discussion](https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384)
15
- * Later batch made using .imatrix file from [this](https://huggingface.co/datasets/ikawrakow/imatrix-from-wiki-train) repo (special thanks to [ikawrakow](https://huggingface.co/ikawrakow) again)
16
 
 
 
 
 
 
17
 
18
  For a brief rundown of iMatrix quant performance please see this [PR](https://github.com/ggerganov/llama.cpp/pull/5747)
19
 
 
11
 
12
  <b>Highly requested model.</b> Quantized from fp16 with love.
13
 
14
+ * <s>1st batch (IQ3_S, IQ3_XS) use a imatrix.dat file calculated from Q8 quant using an input file from [this discussion](https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384)</s> These have been removed in favor of later method. Please see tables below.
15
+ * Later files made using .imatrix file from [this](https://huggingface.co/datasets/ikawrakow/imatrix-from-wiki-train) repo (special thanks to [ikawrakow](https://huggingface.co/ikawrakow) again)
16
 
17
+ <h3>Original IQ3_XS KL Divergence (calculated vs Q8)</h3>
18
+ <img src="https://huggingface.co/InferenceIllusionist/maid-yuzu-v8-alter-iMat-GGUF/resolve/main/IQ3_XS_OLD_KL.JPG?download=true" width="300"/>
19
+
20
+ <h3>Updated IQ3_XS KL Divergence (calculated vs Q8)</h3>
21
+ <img src="https://huggingface.co/InferenceIllusionist/maid-yuzu-v8-alter-iMat-GGUF/resolve/main/IQ3_XS_NEW_KL.JPG?download=true" width="300"/>
22
 
23
  For a brief rundown of iMatrix quant performance please see this [PR](https://github.com/ggerganov/llama.cpp/pull/5747)
24